Chapter Nine: Data Transmission

Chapter Nine:Data Transmission

Introduction• Binary data is transmitted by either by serial or parallel methods

• Data transmission over long distances is usually done by serial means

Data Coding• Data coding techniques are used to convert

alphanumeric characters into binary numbers• A data code is a standardized relationship between

signaling elements and characters• Data codes are also called character sets or

character codes

Binary and Text Data• Alphanumeric characters are not the only source of

data• Digitized audio and video signals are examples of

binary data that has nothing to do with the alphabet• Character codes are not needed for binary data;

however, some communication systems divide all data into blocks that typically contain 7 or 8 bits, referred to as characters

Character Codes• Baudot code dates from 1874 and is still in use in telex and

radioteletype services• It is designated as ITA2 by the CCITT• Baudot is a five-bit code giving a total of 32 possible

combinations. Additional combinations are added by using two characters called LTRS and FIGS

• LTRS includes only uppercase letters• The FIGS character set also includes control characters

ASCII Code• The American Standard Code for Information Interchange (ASCII) is

the most common code for communications between microcomputers• ASCII is a seven-bit code, allowing for 128 possible combinations

without shifting. Sometimes an eighth bit is added to allow for the transmission of graphic characters, mathematical symbols, and foreign language characters

• ASCII has certain regularities that make programming easier; for example, bit 6 switches between upper- and lowercase letters and numbers are represented by their BCD (binary-coded decimal) equivalents

Asynchronous Transmission• Digital transmission formats must have the following

elements:– Standard voltage ranges– Standard clock speeds– Transmitter and receiver clocks must be synchronized– Framing is necessary to determine the first bit of a character or

PCM sample• Asynchronous systems use free-running clocks on the

transmitting and receiving sides• A start bit is transmitted at the beginning of each character

and at least one stop bit at the end of the character

Asynchronous Communication

Parallel-to-Serial Conversion• Internally, computers use parallel communication sending

at least eight and usually 16 or 32 bits along on a bus• Most data communication over longer distances uses serial

techniques• Therefore, a device is needed to convert parallel

signals/data into serial signals/data• Typically, this device is an IC called a UART (universal

asynchronous receiver-transmitter)

UART

Synchronous Transmission• In synchronous systems, the receive and transmitter clocks

are synchronized to the same clock frequency• Synchronous communication is more efficient than

asynchronous because start and stop bits are not necessary• Data is sent in blocks of specific data segments and have

an initial identifying sequence that allows proper framing and often identifies the content of the block as well

Clock Recovery• Asynchronous digital systems can function with a receiver

clock whose speed is only approximately the same as the transmitter clock because the clock is reset at the start of each character

• The most economical way for the receiver to get clock-synchronizing information from the incoming data stream

Framing• Framing is necessary in data communications in order to

determine which bit is the start bit of a character• In asynchronous systems, the use of a start and a stop bit

solves this problem• In synchronous communications, it is only necessary to

determine the start of a block of data

Synchronous Data-Link Protocols• Synchronous communication protocols are either bit- or character-

oriented, with character-oriented protocols becoming obsolete• One of the most common protocols is the BISYNC protocol from IBM• These protocols use at least two synchronizing characters at the

beginning of a block

Error Detection and Correction• Errors will occur with data whether sent by analog or

digital means. Some method must be used to detect and correct these errors

• Errors may be caused by:– High data rates– Thermal noise– RF interference

• There are two stages in the control of errors:– Error detection– Error correction

Techniques for Detection and Correction of Errors

• Some systems rely on the receiver being able to correct errors from redundant information

• This technique is called forward error correction (FEC)• Systems that are full- or half-duplex that request the

transmitter to repeat any data blocks that have errors are called automatic request for retransmission (ARQ) schemes

Parity• Parity is a simple means of error detection by using the

addition of one extra bit to the bits that encode a character• The parity bit is set to make the transmitted character have

either an even number or an odd number of ones• The parity is checked by the receiver and a request is made

to retransmit the character

Longitudinal Redundancy Check• The longitudinal redundancy check (LRC) is an extension

of the parity check that can provide some correction as well as detection

• A parity character is made up from the block of data• By checking the parity of each character, the receiver can

go to the LRC character to determine which bit is in error, allowing the correction of single-bit errors

Checksums• Another method for error detection is the use of checksums• Checksums are developed by adding all the data words in a

block and then dividing by some fixed number• The remainder that results is transmitted at the end of the

block; the receiver performs the same division, and if the numbers are different, an error has occurred

• XMODEM is one data-transfer method that uses this method

Cyclic Redundancy Checking• Cyclic redundancy checking (CRC) codes are used in

many sophisticated digital systems such as the compact disk

• This error-detecting and correcting scheme uses a form of feedback in which the state of each message bit depends on the state of the previous blocks in the message

• CRC codes are generated by computing a polynomial from the message bits

Data Compression and Encryption• Data Compression can be done in several ways, depending

upon the type of data to be processed• Huffman coding uses strings of from 4 to 11 bits to encode

the alphabet instead of 7 or 8 bits for every character• Another important compression technique is called run-

length encoding, taking advantage of the fact that information bits are often repeated

Encryption• Encryption is used to provide security for data

transmissions• Two main types of encryption are:

– Private key - a long binary number is combined with the message data according to an algorithm that is built into the system. At the receiver, the same key is used to decode the information

– Public key - anyone can encode messages using a public key, but a private key is necessary to decode the information